112 
Triassic Echinoderms of Bakony. 
tuberculate, being in fact their ancestors. This, however, is not the case. 
There are some early unituberculate species, to which further reference will 
be made under the name Mesodiadema ; but the pauci-tuberculate Hemipedinas 
seem to have originated in Bajocian time, whereas multituberculate Diadem- 
opsis are common in the Lias. We have therefore to suppose either a ret- 
rogression from Diademopsis to Hemipedina, which is possible but not probable, 
or a subsequent and independent evolution of the typical Bajocian Hemipedina. 
Some of the later multituberculate species ( Phymopedina &c ) may be a further 
development of Hemipedina, or merely a slight modification of Diademopsis. 
On any of these hypotheses Hemipedina must be regarded as starting a distinct 
line of descent. 
General principles therefore lead to the acceptance of the two genera 
Diademopsis and Hemipedina. At the same time they render the construction 
of satisfactory diagnoses a still more difficult task. It seems quite probable 
that II. Woodwardi and H. microgramma have rightly been regarded as 
arising within the Hemipedina line; but rigid adherence to Lambert’s diagnoses 
would place them in Diademopsis. So far I have been unable to lay hold of 
any point by which a true Liassic Diademopsis may be distinguished from a 
multituberculate descendant of a Bajocian Hemipedina. The relations of the 
tubercular series are the same, and although there are differences between 
different species in regard to the composition of the ambulacra, those differ¬ 
ences cannot be groiiped in correlation with the tuberculation. 
In a recent paper (Bull. Soc. Sei. Yonne, 1905; 1906) Dom Aurelien 
Valette has attempted to separate these two genera according to the width 
of the interambulacral plates. Presumably the measurements are to be made 
at the ambitus, but he does not state this. Thus, in Diademopsis the width 
of an interambulacral is more than twice its height, while in Hemipedina it 
is equal to or less than twice the height. Greater width obviously is correlated 
with increased number of tubercle-series, so that Diademopsis is further said 
to have at least two secondary series to each column, with the tubercles not 
much smaller than those of the main series, while Hemipedina has a large 
main series with relatively small secondary series. It does not appear, then, 
that this emphasising of relative width adds anything of importance to the 
diagnosis: it provides no really fresh character by which one can check the 
alternative hypotheses as to the ancestry of such species as H. tetragramma 
and H. marchamensis. Dom Aurelien, however, also maintains that in all 
species referred by him to Diademopsis the ambulacrals are fully formed 
majors of three elements, whereas in Hemipedina the adapical region of the 
ambulacrum has sometimes simple primaries, sometimes incompletely formed 
majors with three granuliferous elements, and sometimes fully formed majors 
each with a perforate tubercle. Unfortunately the author’s division of the 
species is not consistent with the facts in many cases where 1 have been 
able to check it by examination of the type-specimens; therefore it does not 
help the present discussion. He is, however, justified in pointing out that many 
species from the Lias and Infralias have more highly developed ambulacra 
than have several species from later rocks. In this respect also, then, there 
must have been inequality of development, if not actual divergence. lhe 
